Field of the Invention
The present invention relates to pattern recognition and more particularly to home automation utilizing pattern recognition of occupant behavior.
Description of the Related Art
Pattern recognition is a computational algorithm used to classify raw, observed data. The term “pattern recognition stems from machine learning, but has been adapted by cognitive psychologists to describe various theories for how the human brain progresses from incoming sensory information to action selection. The various stages of pattern recognition include the sensing of an environment to detect source data, feature detection in the incoming data, data classification based upon features, and action selection based upon classification. Pattern recognition algorithms have been applied to the behavior of individuals and also to the movements of individuals and vehicles in order to enable predictive organizational systems—in particular traffic flow and management systems and in animal tracking studies.
In a traffic flow and management system, the movements of vehicles are observed over time in order to predict the route and timing of departure of different vehicles. Known formally as travel pattern analysis, the route and timing of departure of different vehicles can be tracked so that traffic patterns for entire communities can be predicted, road construction requirements planed, and traffic lights timed accordingly. In comparison, pattern analysis as applied to the movements of animals has be used scientifically to predict migratory patterns of animals in the wild.
Qualitatively, pattern analysis can be used to identify an individual. In this regard, when an individual repeats the same pattern of movement over time, and such movement is documented as a pattern, a person subsequently engaging in the same pattern of movement can be presumed to be the individual. Pattern analysis of the movement of individuals also finds application in homeland security with particular movements being associated with suspicious behavior. Notwithstanding, pattern matching appears not to have been widely utilized in the implementation of a smart home.